Wrong mass error displayed on top of the spectrum

Rappsilber-Laboratory / xiview

GNU General Public License v2.0

4 stars 1 forks source link

Wrong mass error displayed on top of the spectrum #564

Open WeiqiangChen opened 2 months ago

WeiqiangChen commented 2 months ago

The error on top of the spectrum is 154 Da, while below in the table, it is -0.95ppm. The settings were 6ppm MS1 and 15ppm MS2

https://xiview.org/network.php?upload=26642-44225-82650-47218-81521

colin-combe commented 2 months ago

Hi, yes, there is something strange going on.

The numbers along the top of the Spectrum View are returned from the annotator (which provides the annotations of peaks in the spectra).

I can see the calc M/z in the spectrum view and the table are different, and hence the differing mass errors.

It seems like the calc m/z returned by the annotator is different if one of the peptides contains the (-28.006128) modification.

We'll look into it further, best wishes, Colin

(@lutzfischer - any ideas? the data is from an mzIdentML file)

WeiqiangChen commented 2 months ago

This mzid file was manually modified in Notepad++ to remove all empty cvParam lines, to be able to load into (XiView Issue #563).

Maybe this somehow changes the behavior of XiView.

colin-combe commented 2 months ago

you may be right - e.g. there is missing modification info because of the missing CV term. These empty elements were in Peptide/Modification in the mzIdentML file?

WeiqiangChen commented 2 months ago

Yes. .mzid

colin-combe commented 2 months ago

actaully, i think the masses are not in cv params...

lutzfischer commented 2 months ago

@WeiqiangChen could you upload/email me the config. I am trying and failing to get to the correct mass based on the peptide masses and crosslinker mass (-28Da). Looks like somewhere something got lost but I am not sure what got lost where.

WeiqiangChen commented 2 months ago

ID_24036_R498_with_Phospho_STY.zip The configue file is in the zip

lutzfischer commented 1 month ago

Ok not sure how this actually happened but to me it looks like the modification mentioned in xiVIEW/mzIdentML is not correct. Changing the modification to the configured photo modification (126Da) the masses fit again. Not sure where the -28 comes from here. I guess the crosslinker modification got somehow duplicated and replaced the "photo" (126.something) modification

lutzfischer commented 1 month ago

if you could find the original and edited parts of the mzIdentML that define that peptide pair could be helpful.